PROJECT SUMMARY Pathophysiological changes in Alzheimer's disease (AD) precede clinical manifestations by several decades. Accordingly, one of the priorities of the National Alzheimer's Plan Act is to accelerate efforts to identify the presymptomatic stages of AD by using in-vivo biomarkers. Neuropsychiatric symptoms (NPS) such as depression and anxiety are common in the elderly. The key research question of the field of neuropsychiatry of presymptomatic Alzheimer's disease (AD) is to determine as to which came first, i.e., neuropsychiatric symptoms (NPS) or pathophysiological brain changes related to AD. Such a question can only be answered by using two critically important resources: 1) a large scale, preferably population-based, neuroimaging cohort, and 2) a research infrastructure that can clearly document the lifelong pattern of NPS among participants in the neuroimaging cohort. Building both resources from scratch can be prohibitively expensive and may take decades. In our proposed 4- years R01 grant, we will conduct an in-depth investigation of the pathways linking neuroimaging bi- omarkers, NPS and cognitive outcomes by using the extensive neuroimaging biomarker resource of the Mayo Clinic Study of Aging (MCSA) and we will link this with longitudinal psychiatric data by using the unique resource of the Rochester Epidemiology Project (REP). We hypothesize that the asso- ciation between ?-amyloid deposition, neurodegeneration and the outcome of incident MCI/ AD dementia or trajectories of cognitive changes, is modified by NPS. We will test this hy- pothesis by examining the pathways linking NPS and AD biomarkers in a cohort of over 2,000 cognitively normal persons aged ? 60 years that have undergone amyloid PET, FDG-PET, and MRI of the brain and psychiatric assessment at baseline with at least one follow-up event. The imaging modalities of the proposed R01 are assessed by MCSA. We will measure NPS by using two approaches: 1) Our access to REP medical record linkage system will enable us to rigorously screen for life-time NPS of each study participant. We define NPS as depressive and anxiety symptoms which will be the focus of the primary analysis. The secondary analysis will include other NPS such as apathy and agitation. REP is perhaps the only resource in the world that captures medical data from birth to death on residents of Olmsted County; 2) We will also utilize the assessments used by MCSA (e.g., Beck Depression Inventory and Beck Anxiety Inventory) to augment the REP data. The dependent variables will be trajectories of cognitive changes as continuous outcomes, as well as cat- egorical outcomes of incident MCI/ AD dementia. In conclusion, here we propose a 4-years R01 study that will eventually address the time-honored knowledge gap on time sequence between psy- chiatric symptoms, AD neuroimaging biomarkers and the outcome of incident MCI/ AD dementia.